metadata
license: mit
base_model: Edmon02/speecht5_finetuned_hy
tags:
- generated_from_trainer
model-index:
- name: speecht5_finetuned_voxpopuli_hy
results: []
speecht5_finetuned_voxpopuli_hy
This model is a fine-tuned version of Edmon02/speecht5_finetuned_hy on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6214
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 4
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7717 | 4.4199 | 500 | 0.7195 |
| 0.7361 | 8.8398 | 1000 | 0.6810 |
| 0.7088 | 13.2597 | 1500 | 0.6597 |
| 0.6942 | 17.6796 | 2000 | 0.6465 |
| 0.6835 | 22.0994 | 2500 | 0.6367 |
| 0.6753 | 26.5193 | 3000 | 0.6309 |
| 0.6734 | 30.9392 | 3500 | 0.6245 |
| 0.6747 | 35.3591 | 4000 | 0.6214 |
Framework versions
- Transformers 4.43.3
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1